=?Big5?B?pl7C0DogUmU6IFJlaXNlcmZzIHZzIFhGUyAuLiBOb3cgOiBIb3cg?= =?Big5?B?dG8gaW1wcm92ZSBJTyBkaXNrIHNwZWVkLg==?=

Johan De Meersman jdm at operamail.com
Wed Apr 12 08:36:41 CDT 2006


mcclnx mcc wrote:

> What you say may be true, but technology improve so
> fast where can you buy 4GB, 9 GB disk now.

Those are hard to obtain now, true, but any decent supplier of server
equipment will have ample stock of 36G disks, not to mention 72G ones.

Now, JF has 14 disks of 300G in raid 5, so about 3.9T of storage
assuming N-1 capacity. If you were take it to the extreme, you would do
that with 36G disks and you'd need about 110 disks.

Thus, you could do 110 concurrent seeks instead of 14, read or write 110
blocks concurrently instead of 14, et cetera.

All the above values are of course theoretical, and strongly dependent
on the usage pattern, but the general priciple is there: more disks =
more performance, for the very simple reason that you've got more
independend read/write heads.


If you're talking large (huge) files, on the other hand, Tino is quite
right - disks with greater data density, especially bit density, are
able to read far more data in a given timeframe and at a constant
spinrate when reading sequentially.


-- 
Do not overtax your powers.
-- 

Public GPG key at blackhole.pca.dfn.de

GCS/IT d- s:+ a- C(+++)$ UL++++$ P+++(++++)$ L++(+++)$ !E- W+(+++)$
N+(++) o K w$ !O !M V PS(++)@ PE-(++)@ Y+ PGP++(+++) t(+) 5 X R tv--
b++(++++) DI++(++++) D++ G e++>+++++ h(+) r y+**
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: OpenPGP digital signature
Url : http://lists.us.dell.com/pipermail/linux-poweredge/attachments/20060412/36c44d68/attachment.bin 


More information about the Linux-PowerEdge mailing list